The effects of many drugs decrease over the course of exposure to the drug, the phenomenon being termed "tolerance." One interpretation of tolerance emphasizes the interaction between learning and pharmacology; the organism learns, via Pavlovian conditioning, an association between predrug cues and the systemic effect of the drug. It has frequently been noted that conditional drug responses (elicited by presenting drug- predictive cues without the drug) are anticipatory compensations for the drug effect. According to the conditioning analysis, these conditional drug-compensatory responses contribute to tolerance by attenuating the effect of the drug. It has been hypothesized that drug-compensatory CRs, which mediate tolerance when the drug is administered in the drug-associated environment, may be expressed as "withdrawal symptoms" when the usual drug is not administered in the presence the pre-drug cues. The aims of the proposed experiments are to extend the conditioning analysis to two new areas: (1) intra-administration pharmacological associations and tolerance [i.e., does the early effect of a drug act as a signal for a later effect?], and (2) the associative bases for differences between self-administered and passively received drugs [i.e., why do organisms that self- administer a drug display more tolerance and withdrawal symptoms than do organisms that passively receive the drug?]. The results of the proposed experiments will further our understanding of the role of learning in failures of tolerance (enigmatic overdoses). The results will also be useful in designing learning-based treatment strategies for drug dependence (so-called "cue- exposure" treatment).